"""Random baseline agent for the hospital environment."""
from __future__ import annotations

from typing import Optional

import numpy as np

from hospital_env import HospitalEnv


class RandomAgent:
    """Uniformly samples an action from ``Discrete(num_actions)``.

    Intended purely as a baseline — serves as the lower bound against
    which the heuristic and PPO agents are compared in the grader.
    """

    def __init__(
        self,
        num_actions: int = HospitalEnv.NUM_ACTIONS,
        seed: Optional[int] = None,
    ) -> None:
        self.num_actions = int(num_actions)
        self.rng = np.random.default_rng(seed)

    def act(self, obs: dict) -> int:
        """Return a uniformly random action in ``[0, num_actions)``."""
        return int(self.rng.integers(0, self.num_actions))

    def __call__(self, obs: dict) -> int:
        return self.act(obs)

    def reset(self, seed: Optional[int] = None) -> None:
        """Re-seed the underlying RNG."""
        self.rng = np.random.default_rng(seed)


def _main() -> None:
    """CLI entry point: grade a :class:`RandomAgent` and dump JSON."""
    import json

    from grader import Grader

    agent = RandomAgent(seed=0)
    grader = Grader(n_episodes_per_scenario=3)
    scores = grader.grade(agent)
    print(json.dumps(scores, indent=2))


if __name__ == "__main__":
    _main()